Gene fusions have been considered a hallmark of neoplasia. The study of gene fusions has founded the theoretical backgrounds for many cancer diagnosis and therapeutics. Recently, we and others reported a chimeric fusion RNA involving two neighboring genes, SLC45A3 and ELK4, in prostate cancer. The chimeric RNA expression level correlates with prostate cancer progression. In addition, silencing the chimera led to slower proliferation and higher levels of CDKN1A expression in both androgen-dependent and castration- resistant prostate cancer cells. Intriguingly, we found that SLC45A3-ELK4 is generated by a mechanism of cis- splicing of adjacent genes (cis-SAGe)/read-through, instead of chromosomal rearrangement. Despite the biological and clinical significance of SLC45A3-ELK4, neither has the mechanism of cis-SAGe been elucidated, nor have other examples of cis-SAGe been identified. In our preliminary studies we hypothesized, and have accumulated multiple lines of evidence, that the binding of the zinc finger protein CTCF to insulator sites in-between the two neighboring genes plays an important role in the generation of cis-SAGe between SLC45A3 and ELK4. We also hypothesize that more cis-SAGe events can be identified through manipulating CTCF level, with implications for both prostate cancer biology and tumor biology in general. We propose to provide more evidence for these hypotheses in three aims. In Aim1, we will investigate the mechanism of cis- SAGe using SLC45A3-ELK4 as a model. We reasoned that the cis-SAGe is essentially alternative splicing of a continuous transcript passing through gene boundaries. We hypothesized that three factors have to be met for the cis-SAGe event to happen between SLC45A3 and ELK4: activation of the 5' SLC45A3 gene, reduced CTCF binding to the insulator regions in-between two genes, and enhanced alternative splicing. In Aim1, we will investigate all three factors of regulation first in LNCaP prostate cancer cell line, then in other cell lines and in clinical samples; In Aim2, we will manipulate CTCF level and use paired-end transcriptome sequencing to identify additional cis-SAGe fusions. Our preliminary study using LNCaP cells already generated many fusion candidates for study. We will further validate these candidates as true cis-SAGe events and expand to other cell lines. In Aim3, we will investigate the biological and clinical significance o the newly identified cis-SAGe fusions. The proposed study will shed light on the mechanism of cis-SAGe and discover novel cancer- associated chimeric transcripts. The novel fusion gene products may be potential candidates for new disease biomarkers and/or therapeutic targets. The same approach can be easily adapted in other cancer types.